A5110516236- Spectroscopy and Chemometric Analyses
- Smart Agriculture and AI
- Data Management and Algorithms
- Image Processing Techniques and Applications
- Remote Sensing in Agriculture
- Advanced Algorithms and Applications
- Animal Behavior and Welfare Studies
- Constraint Satisfaction and Optimization
- Impact of Light on Environment and Health
- Face and Expression Recognition
- Advanced Clustering Algorithms Research
- Remote-Sensing Image Classification
- Semantic Web and Ontologies
- Hand Gesture Recognition Systems
- Fault Detection and Control Systems
- Optical Coherence Tomography Applications
- Human Mobility and Location-Based Analysis
- Complex Network Analysis Techniques
- Industrial Vision Systems and Defect Detection
- Geographic Information Systems Studies
- Remote Sensing and Land Use
- Human Pose and Action Recognition
- Effects of Environmental Stressors on Livestock
- Remote Sensing and LiDAR Applications
- Leaf Properties and Growth Measurement
Ministry of Natural Resources
2024
Jilin Agricultural University
2010-2024
Jilin University
2013
Jilin Agricultural Science and Technology University
2013
Jilin Province Science and Technology Department
2013
The number of wheat spikes per unit area is one the most important agronomic traits associated with yield. However, quick and accurate detection for counting faces persistent challenges due to complexity field conditions. This work has trained a RetinaNet (SpikeRetinaNet) based on several optimizations detect count efficiently. consists improvements. First, weighted bidirectional feature pyramid network (BiFPN) was introduced into (FPN) RetinaNet, which could fuse multiscale features...
In complex citrus orchard environments, light changes, branch shading, and fruit overlapping impact detection accuracy. This paper proposes the model YOLO-DCA in environments based on YOLOv7-tiny model. We used depth-separable convolution (DWConv) to replace ordinary ELAN, which reduces number of parameters model; we embedded coordinate attention (CA) into make it a (CAConv) neck network convolution; dynamic head original head. trained evaluated test using homemade dataset. The size is 4.5...
Wheat is a crucial crop worldwide, and accurate detection counting of wheat spikes are vital for yield estimation breeding. However, these tasks daunting in complex field environments. To tackle this, we introduce RIA-SpikeNet, model designed to detect count such conditions. First, an Implicit Decoupling Detection Head incorporate more implicit knowledge, enabling the better distinguish visually similar spikes. Second, Asymmetric Loss employed as confidence loss function, enhancing learning...
The image segmentation of corn diseases is one the critical technical aspects digital processing technology for disease recognition. Currently, aiming at solving loss colure texture segmentation, this paper proposes a modified bionic pulse coupled neural network. This proposed algorithm combined from network and artificial bee colony algorithm. A revenue function defined based on linear weighted with maximum Shannon entropy minimum cross-entropy. Through adaptive strategy searching...
Because of using the possibilistic partition to describe data set, clustering algorithm is more robust noises than hard and fuzzy algorithms. But calculating membership matrix also makes it has a low efficiency. Moreover, performance may be degreased if cluster number set wrongly. In this paper, we proposed new named unsupervised c-means with reduction (UPCMDR) improve efficiency (PCM). UPCMDR, technique introduced speed up process estimation centers. A called weighted estimate positions...
The classification of urban functional zones is crucial for improving land use efficiency and promoting balanced development across areas. Existing methods classifying using mobile signaling data face challenges primarily due to the limitations single sources, insufficient utilization multidimensional data, inherent inaccuracies in data. To address these issues, this study proposes an innovative method that employs advanced multimodal fusion techniques enhance accuracy reliability zone...
In the on-line monitoring for fault of rolling bearing, we have no information about cluster number obtained data signal, which cause great challenges diagnosis when using clustering algorithms. this paper, extract three features vibration signals bearings as parameters in time-domain, and then multi-scale possibilistic (MPCM) algorithm is applied to unsupervised diagnosis. Because clusters controlled by factors automatically, proposed method detection can meet actual demand machinery...
There are representative models for spatial topological relations of Region connection calculus and intersections model. The purpose this paper is to study the multiple simple regions. In paper, 4-intersection matrix model extended into a 2n-intersection function represent exclusiveness completeness these proved. Moreover, also gives detailed discussion on properties applications from special new perspective. And among three regions in applied forecast impact typhoon specified target areas....
Since the beginning of 21th century, with continuous development information technology, it has been difficulty and hotpot in understanding field to study analyze sports videos computer technology. Because soccer have a broad mass background, is more valuable research on it. At present, there are many researches based trajectory players or balls. But due background interference other reasons, detection accuracy still needs be improved. In this paper, we present method tactical behavior...
The manual intervention before the birth dates is very important to reduce dystocia rate and enhance survival of calves effectively. Information acquisition cows' basic sow-activity modes regulars one judgments critical points. Now, mainstream way has limitations that external sensors are used obtain information by attaching specific positions. So, we proposed a method for recognizing sow-activities using bag visual words under video monitoring in this paper. Firstly, detected remarkable...